This thread has been locked.

If you have a related question, please click the "Ask a related question" button in the top right corner. The newly created question will be automatically linked to this question.

OPA1622: Simulation problems with cascaded outputstage with OPA1622

Part Number: OPA1622
Other Parts Discussed in Thread: OPA1611, OPA2192

Hi,

I am currently working on a headphone amplifier. The goal of the headphone amplifier is to have unmeasurably low distortion levels. For this reason a cascaded outputstage with global feedback will be used to increase loopgain and therefor limit the disturbances introduced to the system.

The parts I used are the following:
- OPA1622 (used for output stage).
- OPA1611/OPA1622 (used in buffer stages and in the cascaded output stage).
- OPA2192 (used to remove DC offset).

Since global feedback is used, stability can become a problem (especially at high frequencies). For that reason I tried doing a stability analysis. I did this by modelling the parts in MATLAB. The modelling was entirely done based off the gain/phase plots.

My model of the OPA1622 looks as follows:
Zeros: [-15119038.5932601]
Poles: [-135.367123896864 -3319616.85982940]

Gain: [-62291218.1241616]

This resulted in the following bode plot:

 

The same approach was done for the OPA1611 and led to the following result:
Zeros: [-2.2796e+08]
Poles: [-1141.4 -5.8245e+07 -2.8662e+09]
Gain:[ -4.6737e+17]


Cascading them in unity gain leads to some pretty wonky transients. For that reason and for band limiting I decided to add filter functionality by adding two poles at 100kHz with a damping of 1 (Q=0.5).

This lead to some fantastic results:


And working as intended.

The poles are placed by configuring the cascaded OPA1611 and OPA1622 with a multiple feedback filter (http://sim.okawa-denshi.jp/en/OPttool.php). The reason for this is because it’s very easy to convert it to a differential filter (which will be done).

The simulation right now is in ideal circumstances. I want to verify it with the models provided by TI. And thus I proceeded to do in TINA TI. The input buffers are working as intended, my offset removal circuitry works as intended (2nd order low pass that I later subtract from the original signal). The only thing that doesn’t exactly work is my fabulous outputstage.

I don’t seem to get any errors with the transient analysis. With the AC analysis I tend to get the error “Operating point not found”.
How can I solve there errors and proceed with my simulations?

I can share my design in private, but I prefer to not keep it out in the open on the internet.

Kind regards,

Hidde